Developers can request storage using persistent volume claims.

Create PersistentVolumeClaim and a MySQL pod which references the fast StorageClass:

kubectl create -f storageos-pvc.yaml

kubectl create -f mysql-pod1.yaml

You should see a StorageOS persistent volume:

kubectl describe pv

Usually takes ~30 seconds to come up after it's been created.

You can also confirm StorageOS provisioned the volume with the CLI:

storageos volume ls --format "table {{.Name}}\t{{.Size}}\t{{.Replicas}}\t{{.Location}}"

The link between the PersistentVolumeClaim and the pod comes from the pod referencing the PersistentVolumeClaim in the pod's yaml definition.

Run the command to see the mysql-pod definition.

cat mysql-pod1.yaml

You can see that the PersistentVolumeClaim claimName has been set to mysqlpvc.

Run the command to see the PersistentVolumeClaim definition.

cat storageos-pvc.yaml

Notice that the PersistentVolumeClaim name is mysqlpvc.

Check the pods are in RUNNING state:

kubectl get pods -w

The above command watches the pod created in the default namespace. Press Ctrl+C to continue once the pods are up.

Connect to MySQL and create a test database (your prompt will change):

kubectl exec -it mysql -- mysql

CREATE DATABASE shop;

Now connect to the database and insert some sample data:

USE shop;

CREATE TABLE FRUIT( ID INT PRIMARY KEY NOT NULL, INVENTORY VARCHAR(25) NOT NULL, QUANTITY INT NOT NULL );

INSERT INTO FRUIT (ID,INVENTORY,QUANTITY) VALUES (1, 'Bananas', 132), (2, 'Apples', 165), (3, 'Oranges', 219);

Check your commands succeeded:

SELECT * FROM FRUIT;

Quit the MySQL container:

\q

Now, let's check that the data was persisted and is accessible on another node.

Make a note of the location on which the volume got initially prosioned on:

storageos volume ls --format "table {{.Name}}\t{{.Size}}\t{{.Replicas}}\t{{.MountedBy}}\t{{.Location}}"

Save the namespace/volume into a variable.

volume=$(storageos volume ls --format "table {{.Name}}\t{{.Replicas}}" | grep '0/0' | awk '{ print $1 }')

Create 1 replica for the volume, by adding a label to the volume.

storageos volume update --label-add storageos.com/replicas=1 $volume

Check that the replica was created.

storageos volume ls --format "table {{.Name}}\t{{.Replicas}}"

Usually takes ~10 seconds to create the replica in this example.

In order to keep this tutorial simple, we are going to delete the MySQL pod and create it on a different node in our cluster.

kubectl delete pod mysql

Now, you can see the mount was removed but the location is still the same.

storageos volume ls --format "table {{.Name}}\t{{.Size}}\t{{.Replicas}}\t{{.MountedBy}}\t{{.Location}}"

Create the MySQL pod again but on a different node this time.

kubectl create -f mysql-pod2.yaml

kubectl get pods -w

The above command watches the pod created in the default namespace. Press Ctrl+C to continue once the pods are up.

Connect to the database.

kubectl exec -it mysql -- mysql

USE shop;

Insert more data, then check that there are four records in the table.

INSERT INTO FRUIT (ID,INVENTORY,QUANTITY) VALUES (4, 'Peaches', 203);

SELECT * FROM FRUIT;

Quit the MySQL container.

\q

If you check the location of the volume you will see it has not changed, despite the volume now being mounted by a different host.

storageos volume ls --format "table {{.Name}}\t{{.Size}}\t{{.Status}}\t{{.MountedBy}}\t{{.Location}}"

In this way you can see that the location of the volume is transparent to the pod that mounts the volume.

Congratulations, you have successfully installed StorageOS on Kubernetes, created a volume, and bound that volume to an application!